Exploring The Techniques and Applications of Hand Gesture Recognition for Human-Computer Interaction

Abstract

Hand gesture controlled applications are software applications that are controlled by the user's hand movements. Hand gesture controlled applications have the potential to improve user experience by offering more intuitive and immersive ways to interact with system. Hand gesture controlled applications have a huge scope, including gaming, virtual reality, robotics, healthcare, and various other sectors. In this paper, we have combined several disparate end results of subsystems such as virtual mouse, contactless keyboard, finger counting, taking selfies, sign language, game, handling installed applications and volume control using computer vision. For implementation, we used the open-source software library OpenCV, mediapipe and pyautogui. Various hand gestures are assigned to launch and control different applications. These hand gestures are recognized using machine learning trained models, Euclidean distance formula and pixel positions. Overall, it allows humans to interact with computers through hand gestures rather than traditional input/output mechanisms such as the keyboard and mouse, improving accessibility. Key Words: Computer vision, OpenCV, hand gesture, machine learning, TensorFlow, pyautogui, mediapipe, virtual mouse, virtual keyboard, sign language, gaming.